Use gitbook-html as base parser

16 files changed, 158 insertions, 396 deletions

diff --git a/packages/gitbook-markdown/lib/glossary.js b/packages/gitbook-markdown/lib/glossary.js
deleted file mode 100644
index 256b52c..0000000
--- a/packages/gitbook-markdown/lib/glossary.js
+++ /dev/null
@@ -1,52 +0,0 @@
-var _ = require('lodash');
-var kramed = require('kramed');
-
-// Get all the pairs of header + paragraph in a list of nodes
-function groups(nodes) {
-    // A list of next nodes
-    var next = nodes.slice(1).concat(null);
-
-    return _.reduce(nodes, function(accu, node, idx) {
-        // Skip
-        if(!(
-            node.type === 'heading' &&
-            (next[idx] && next[idx].type === 'paragraph')
-        )) {
-            return accu;
-        }
-
-        // Add group
-        accu.push([
-            node,
-            next[idx]
-        ]);
-
-        return accu;
-    }, []);
-}
-
-function parseGlossary(src) {
-    var nodes = kramed.lexer(src);
-
-    return groups(nodes)
-    .map(function(pair) {
-        // Simplify each group to a simple object with name/description
-        return {
-            name: pair[0].text,
-            description: pair[1].text,
-        };
-    });
-}
-
-function glossaryToMarkdown(glossary) {
-    var bl = "\n";
-
-    var body = _.map(glossary, function(entry) {
-        return "## "+entry.name+bl+bl+entry.description;
-    }).join(bl+bl);
-
-    return "# Glossary"+bl+bl+body;
-}
-
-module.exports = parseGlossary;
-module.exports.toText = glossaryToMarkdown;
diff --git a/packages/gitbook-markdown/lib/index.js b/packages/gitbook-markdown/lib/index.js
index 55ba205..1f575a5 100644
--- a/packages/gitbook-markdown/lib/index.js
+++ b/packages/gitbook-markdown/lib/index.js
@@ -1,10 +1,8 @@
-var kramed = require("kramed");
-var _ = require("lodash");
+var htmlParser = require('gitbook-html');
 
-module.exports = {
-    summary: require("./summary"),
-    glossary: require("./glossary"),
-    langs: require("./langs"),
-    readme: require("./readme"),
-    page: require("./page")
-};
+var toHTML = require('./tohtml');
+var toMarkdown = require('./tomarkdown');
+var page = require('./page');
+
+module.exports = htmlParser.createParser(toHTML, toMarkdown);
+module.exports.page.prepare = page.prepare;
diff --git a/packages/gitbook-markdown/lib/langs.js b/packages/gitbook-markdown/lib/langs.js
deleted file mode 100644
index 3da906d..0000000
--- a/packages/gitbook-markdown/lib/langs.js
+++ /dev/null
@@ -1,21 +0,0 @@
-var _ = require("lodash");
-var parseEntries = require("./summary").entries;
-
-
-function parseLangs(content) {
-    return parseEntries(content);
-}
-
-function langsToMarkdown(langs) {
-    var bl = "\n";
-    var content = "# Languages"+bl+bl;
-
-    _.each(langs, function(lang) {
-        content = content + "* ["+lang.title+"]("+lang.path+")"+bl;
-    });
-
-    return content;
-}
-
-module.exports = parseLangs;
-module.exports.toText = langsToMarkdown;
diff --git a/packages/gitbook-markdown/lib/page.js b/packages/gitbook-markdown/lib/page.js
index 3a92086..d8a5119 100644
--- a/packages/gitbook-markdown/lib/page.js
+++ b/packages/gitbook-markdown/lib/page.js
@@ -1,10 +1,9 @@
 var _ = require('lodash');
 var kramed = require('kramed');
-
 var annotate = require('kramed/lib/annotate/');
 
-var RAW_START = "{% raw %}";
-var RAW_END = "{% endraw %}";
+var RAW_START = '{% raw %}';
+var RAW_END = '{% endraw %}';
 
 function escape(str) {
     return RAW_START + str + RAW_END;
@@ -15,6 +14,7 @@ function combine(nodes) {
     return _.pluck(nodes, 'raw').join('');
 }
 
+// Add templating "raw" to code blocks
 function preparePage(src) {
     var lexed = annotate.blocks(src);
     var levelRaw = 0;
@@ -54,28 +54,6 @@ function preparePage(src) {
     return combine(escaped);
 }
 
-function parsePage(src) {
-    var options = _.extend({}, kramed.defaults, {
-        mathjax: false,
-        renderer: new kramed.Renderer({
-            langPrefix: 'lang-',
-            smartypants: false,
-            headerPrefix: '',
-            headerAutoId: false,
-            xhtml: false
-        })
-    });
-
-    return {
-        sections: [
-            {
-                type: "normal",
-                content: kramed(src, options)
-            }
-        ]
-    };
-}
-
-// Exports
-module.exports = parsePage;
-module.exports.prepare = preparePage;
+module.exports = {
+    prepare: preparePage
+};
diff --git a/packages/gitbook-markdown/lib/readme.js b/packages/gitbook-markdown/lib/readme.js
deleted file mode 100644
index 9d8f552..0000000
--- a/packages/gitbook-markdown/lib/readme.js
+++ /dev/null
@@ -1,45 +0,0 @@
-var _ = require('lodash');
-var kramed = require('kramed');
-var textRenderer = require('kramed-text-renderer');
-
-function extractFirstNode(nodes, nType) {
-    return _.chain(nodes)
-    .filter(function(node) {
-        return node.type == nType;
-    })
-    .pluck("text")
-    .first()
-    .value();
-}
-
-
-function parseReadme(src) {
-    var nodes, title, description;
-    var renderer = textRenderer();
-
-    // Parse content
-    nodes = kramed.lexer(src);
-
-    title = extractFirstNode(nodes, "heading") || '';
-    description = extractFirstNode(nodes, "paragraph") || '';
-
-    var convert = _.compose(
-        function(text) {
-            return _.unescape(text.replace(/(\r\n|\n|\r)/gm, ""));
-        },
-        function(text) {
-            return kramed.parse(text, _.extend({}, kramed.defaults, {
-                renderer: renderer
-            }));
-        }
-    );
-
-    return {
-        title: convert(title),
-        description: convert(description)
-    };
-}
-
-
-// Exports
-module.exports = parseReadme;
diff --git a/packages/gitbook-markdown/lib/summary.js b/packages/gitbook-markdown/lib/summary.js
deleted file mode 100644
index 77284cf..0000000
--- a/packages/gitbook-markdown/lib/summary.js
+++ /dev/null
@@ -1,179 +0,0 @@
-var _ = require('lodash');
-var kramed = require('kramed');
-
-
-// Utility function for splitting a list into groups
-function splitBy(list, starter, ender) {
-    var starts = 0;
-    var ends = 0;
-    var group = [];
-
-    // Groups
-    return _.reduce(list, function(groups, value) {
-        // Ignore start and end delimiters in resulted groups
-        if(starter(value)) {
-            starts++;
-        } else if(ender(value)) {
-            ends++;
-        }
-
-        // Add current value to group
-        group.push(value);
-
-        // We've got a matching
-        if(starts === ends && starts !== 0) {
-            // Add group to end groups
-            // (remove starter and ender token)
-            groups.push(group.slice(1, -1));
-
-            // Reset group
-            group = [];
-        }
-
-        return groups;
-    }, []);
-}
-
-function skipSpace(nodes) {
-    return _.filter(nodes, function(node) {
-        return node && node.type != 'space';
-    });
-}
-
-function correctLoose(nodes) {
-    return _.map(nodes, function(node) {
-        // Return normal nodes
-        if(!node || node.type != 'loose_item_start') {
-            return node
-        }
-
-        // Correct loose items
-        node.type = 'list_item_start';
-
-        return node;
-    })
-}
-
-function listSplit(nodes, start_type, end_type) {
-    return splitBy(nodes, function(el) {
-        return el.type === start_type;
-    }, function(el) {
-        return el.type === end_type;
-    });
-}
-
-// Get the biggest list
-// out of a list of kramed nodes
-function filterList(nodes) {
-    return _.chain(nodes)
-    .toArray()
-    .dropWhile(function(el) {
-        // Get everything after list_start
-        return el.type !== 'list_start';
-    })
-    .reverse()
-    .dropWhile(function(el) {
-        // Get everything after list_end (remember we're reversed)
-        return el.type !== 'list_end';
-    })
-    .reverse()
-    .value().slice(1, -1);
-}
-
-// Parses an Article or Chapter title
-// supports extracting links
-function parseTitle(src) {
-    // Check if it's a link
-    var matches = kramed.InlineLexer.rules.link.exec(src);
-
-    // Not a link, return plain text
-    if(!matches) {
-        return {
-            title: src,
-            path: null,
-        };
-    }
-
-    return {
-        title: matches[1],
-        path: matches[2],
-    };
-}
-
-function parseChapter(nodes) {
-    var node = _.first(nodes);
-    if (!node) return null;
-    if (!node.text) throw new Error("Invalid entry in the SUMMARY");
-
-    return _.extend(parseTitle(node.text), {
-        articles: _.chain(listSplit(filterList(nodes), 'list_item_start', 'list_item_end'))
-            .map(function(nodes, i) {
-                return parseChapter(nodes);
-            })
-            .compact()
-            .value()
-    });
-}
-
-function listGroups(src) {
-    var nodes = kramed.lexer(src);
-
-    // Get out groups of lists
-    return listSplit(
-        filterList(correctLoose(skipSpace(nodes))),
-        'list_item_start', 'list_item_end'
-    );
-}
-
-function parseSummary(src) {
-    // Split out chapter sections
-    var chapters = _.chain(listGroups(src))
-        .map(parseChapter)
-        .compact()
-        .value();
-
-    return {
-        chapters: chapters
-    };
-}
-
-function parseEntries(src) {
-    return _.chain(listGroups(src))
-        .map(parseChapter)
-        .compact()
-        .value();
-}
-
-
-function summaryToMarkdown(summary) {
-    var bl = "\n";
-    var content = "# Summary"+bl+bl;
-
-    var _base = function(article) {
-        if (article.path) {
-            return "* ["+article.title+"]("+article.path+")";
-        } else {
-            return "* "+article.title;
-        }
-    };
-
-    var convertArticle = function(article, d) {
-        content = content + Array(4*d).join(" ") + _base(article)+bl;
-        _.each(article.articles, function(_article) {
-            convertArticle(_article, d + 1);
-        });
-    };
-
-    _.each(summary.chapters, function(chapter) {
-        convertArticle(chapter, 0);
-    });
-
-    content = content+bl;
-
-    return content;
-};
-
-
-module.exports = parseSummary;
-module.exports.entries = parseEntries;
-module.exports.toText = summaryToMarkdown;
diff --git a/packages/gitbook-markdown/lib/tohtml.js b/packages/gitbook-markdown/lib/tohtml.js
new file mode 100644
index 0000000..9647363
--- /dev/null
+++ b/packages/gitbook-markdown/lib/tohtml.js
@@ -0,0 +1,20 @@
+var _ = require('lodash');
+var kramed = require('kramed');
+
+// Convert markdown to HTML
+function convertMdToHTML(src) {
+     var options = _.extend({}, kramed.defaults, {
+        mathjax: false,
+        renderer: new kramed.Renderer({
+            langPrefix: 'lang-',
+            smartypants: false,
+            headerPrefix: '',
+            headerAutoId: false,
+            xhtml: false
+        })
+    });
+
+    return kramed(src, options);
+}
+
+module.exports = convertMdToHTML;
diff --git a/packages/gitbook-markdown/lib/tomarkdown.js b/packages/gitbook-markdown/lib/tomarkdown.js
new file mode 100644
index 0000000..5bee4e0
--- /dev/null
+++ b/packages/gitbook-markdown/lib/tomarkdown.js
@@ -0,0 +1,43 @@
+
+// Return N time a string
+function ns(s, n) {
+    return Array(n + 1).join(s);
+}
+
+module.exports = {
+    onTitleStart: function(level) {
+        return ns('#', level) + ' ';
+    },
+    onTitleEnd: function(level) {
+        return this.onBL();
+    },
+
+    onParagraphStart: function() {
+        return this.onSection();
+    },
+    onParagraphEnd: function() {
+        return this.onSection();
+    },
+
+    onLinkStart: function() {
+        return '[';
+    },
+    onLinkEnd: function(href) {
+        return '](' + href +')';
+    },
+
+    onListStart: function(level) {
+        return '';
+    },
+    onListEnd: function() {
+        return '';
+    },
+
+    onListItemStart: function(level) {
+        return ns(' ', level * 4) + '* ';
+    },
+    onListItemEnd: function() {
+        return '';
+    },
+};
+
diff --git a/packages/gitbook-markdown/package.json b/packages/gitbook-markdown/package.json
index 97c1df1..d967f26 100644
--- a/packages/gitbook-markdown/package.json
+++ b/packages/gitbook-markdown/package.json
@@ -7,13 +7,14 @@
     "dependencies": {
         "lodash": "^3.2.0",
         "kramed": "0.5.5",
-        "kramed-text-renderer": "0.2.1"
+        "kramed-text-renderer": "0.2.1",
+        "gitbook-html": "1.0.0"
     },
     "devDependencies": {
-        "mocha": "2.3.3"
+        "mocha": "^2.3.3"
     },
     "scripts": {
-        "test": "export TESTING=true; mocha --reporter list"
+        "test": "export TESTING=true; mocha --reporter list --bail"
     },
     "repository": {
         "type": "git",
diff --git a/packages/gitbook-markdown/test/fixtures/GLOSSARY.md b/packages/gitbook-markdown/test/fixtures/GLOSSARY.md
index 5969902..19142d8 100644
--- a/packages/gitbook-markdown/test/fixtures/GLOSSARY.md
+++ b/packages/gitbook-markdown/test/fixtures/GLOSSARY.md
@@ -1,15 +1,15 @@
-# Magic
+## Magic
 Sufficiently advanced technology, beyond the understanding of the observer producing a sense of wonder.
 
 Hello, I am random noise in the middle of this beautiful Glossary. (Really astonishing !)
 
-# PHP
+## PHP
 An atrocious language, invented for the sole purpose of inflicting pain and suffering amongst the proframming wizards of this world.
 
-# Clojure
+## Clojure
 Lisp re-invented for hipsters.
 
-# Go
+## Go
 Go Go Google [Wow](https://www.google.com)
 
 Fantastic, I love code too ! :
@@ -25,6 +25,6 @@ def f(x):
 print(f(9))
 ```
 
-# Gitbook
+## Gitbook
 
 Awesome project. Really amazing, I'm really at a loss for words ...
diff --git a/packages/gitbook-markdown/test/fixtures/SUMMARY_PARTS.md b/packages/gitbook-markdown/test/fixtures/SUMMARY_PARTS.md
new file mode 100644
index 0000000..eaba1a7
--- /dev/null
+++ b/packages/gitbook-markdown/test/fixtures/SUMMARY_PARTS.md
@@ -0,0 +1,11 @@
+# Summary
+
+* [Chapter 1](chapter-1/README.md)
+
+## Part 2
+
+* [Chapter 2](chapter-2/README.md)
+
+## Part 3
+
+* [Chapter 3](chapter-3/README.md)
diff --git a/packages/gitbook-markdown/test/glossary.js b/packages/gitbook-markdown/test/glossary.js
index efa77db..3231020 100644
--- a/packages/gitbook-markdown/test/glossary.js
+++ b/packages/gitbook-markdown/test/glossary.js
@@ -4,10 +4,14 @@ var assert = require('assert');
 
 var glossary = require('../').glossary;
 
-var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/GLOSSARY.md'), 'utf8');
-var LEXED = glossary(CONTENT);
-
 describe('Glossary parsing', function () {
+    var LEXED;
+
+    before(function() {
+        var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/GLOSSARY.md'), 'utf8');
+        LEXED = glossary(CONTENT);
+    });
+
     it('should only get heading + paragraph pairs', function() {
         assert.equal(LEXED.length, 5);
     });
@@ -19,7 +23,7 @@ describe('Glossary parsing', function () {
     });
 
     it('should correctly convert it to text', function() {
-    	var text = glossary.toText(LEXED);
-    	assertObjectsEqual(glossary(text), LEXED);
+        var text = glossary.toText(LEXED);
+        assertObjectsEqual(glossary(text), LEXED);
     });
 });
diff --git a/packages/gitbook-markdown/test/langs.js b/packages/gitbook-markdown/test/langs.js
index e6c3dbd..bb75d11 100644
--- a/packages/gitbook-markdown/test/langs.js
+++ b/packages/gitbook-markdown/test/langs.js
@@ -4,10 +4,14 @@ var assert = require('assert');
 
 var langs = require('../').langs;
 
-var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/LANGS.md'), 'utf8');
-var LEXED = langs(CONTENT);
-
 describe('Languages parsing', function () {
+    var LEXED;
+
+    before(function() {
+        var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/LANGS.md'), 'utf8');
+        LEXED = langs(CONTENT);
+    });
+
     it('should detect paths and titles', function() {
         assert.equal(LEXED.length, 2);
         assert.equal(LEXED[0].path,'en/');
diff --git a/packages/gitbook-markdown/test/page.js b/packages/gitbook-markdown/test/page.js
index 8f4b5c5..8b882d8 100644
--- a/packages/gitbook-markdown/test/page.js
+++ b/packages/gitbook-markdown/test/page.js
@@ -4,25 +4,22 @@ var assert = require('assert');
 
 var page = require('../').page;
 
-function loadPage (name, options) {
-    var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/' + name + '.md'), 'utf8');
-    return page(CONTENT, options).sections;
-}
-
-var LEXED = loadPage('PAGE');
 
 describe('Page parsing', function() {
-    it('should detect sections', function() {
-        assert.equal(LEXED.length, 1);
+    var LEXED;
+
+    before(function() {
+        var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/PAGE.md'), 'utf8');
+        LEXED = page(CONTENT);
     });
 
-    it('should gen content for normal sections', function() {
-        assert(LEXED[0].content);
+    it('should gen content', function() {
+        assert(LEXED.content);
     });
 
     it('should not add id to headings', function() {
-        assert.equal(page('# Hello').sections[0].content, '<h1>Hello</h1>\n');
-         assert.equal(page('# Hello {#test}').sections[0].content, '<h1 id="test">Hello </h1>\n');
+        assert.equal(page('# Hello').content, '<h1>Hello</h1>\n');
+         assert.equal(page('# Hello {#test}').content, '<h1 id="test">Hello </h1>\n');
     });
 
     it('should escape codeblocks in preparation (1)', function() {
diff --git a/packages/gitbook-markdown/test/readme.js b/packages/gitbook-markdown/test/readme.js
index b3a5952..dd059fb 100644
--- a/packages/gitbook-markdown/test/readme.js
+++ b/packages/gitbook-markdown/test/readme.js
@@ -4,11 +4,13 @@ var assert = require('assert');
 
 var readme = require('../').readme;
 
-
-var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/README.md'), 'utf8');
-var LEXED = readme(CONTENT);
-
 describe('Readme parsing', function () {
+    var LEXED;
+
+    before(function() {
+        var CONTENT = fs.readFileSync(path.join(__dirname, './fixtures/README.md'), 'utf8');
+        LEXED = readme(CONTENT);
+    });
 
     it('should contain a title', function() {
         assert(LEXED.title);
diff --git a/packages/gitbook-markdown/test/summary.js b/packages/gitbook-markdown/test/summary.js
index a1bb49a..22800d4 100644
--- a/packages/gitbook-markdown/test/summary.js
+++ b/packages/gitbook-markdown/test/summary.js
@@ -13,56 +13,57 @@ function lex(fixtureFile) {
     );
 }
 
-var LEXED = lex('SUMMARY.md');
-
 describe('Summary parsing', function () {
+    var LEXED;
+
+    before(function() {
+        LEXED = lex('SUMMARY.md');
+        PART = LEXED.parts[0];
+    });
+
     it('should detect chapters', function() {
-        assert.equal(LEXED.chapters.length, 5);
+        assert.equal(PART.articles.length, 5);
     });
 
     it('should support articles', function() {
-        assert.equal(LEXED.chapters[0].articles.length, 2);
-        assert.equal(LEXED.chapters[1].articles.length, 0);
-        assert.equal(LEXED.chapters[2].articles.length, 0);
+        assert.equal(PART.articles[0].articles.length, 2);
+        assert.equal(PART.articles[1].articles.length, 0);
+        assert.equal(PART.articles[2].articles.length, 0);
     });
 
     it('should detect paths and titles', function() {
-        assert(LEXED.chapters[0].path);
-        assert(LEXED.chapters[1].path);
-        assert(LEXED.chapters[2].path);
-        assert(LEXED.chapters[3].path);
-        assert.equal(LEXED.chapters[4].path, null);
+        assert(PART.articles[0].path);
+        assert(PART.articles[1].path);
+        assert(PART.articles[2].path);
+        assert(PART.articles[3].path);
+        assert.equal(PART.articles[4].path, null);
 
-        assert(LEXED.chapters[0].title);
-        assert(LEXED.chapters[1].title);
-        assert(LEXED.chapters[2].title);
-        assert(LEXED.chapters[3].title);
-        assert(LEXED.chapters[4].title);
+        assert(PART.articles[0].title);
+        assert(PART.articles[1].title);
+        assert(PART.articles[2].title);
+        assert(PART.articles[3].title);
+        assert(PART.articles[4].title);
     });
 
     it('should normalize paths from .md', function() {
-        assert.equal(LEXED.chapters[0].path,'chapter-1/README.md');
-        assert.equal(LEXED.chapters[1].path,'chapter-2/README.md');
-        assert.equal(LEXED.chapters[2].path,'chapter-3/README.md');
+        assert.equal(PART.articles[0].path,'chapter-1/README.md');
+        assert.equal(PART.articles[1].path,'chapter-2/README.md');
+        assert.equal(PART.articles[2].path,'chapter-3/README.md');
+    });
+
+    it('should part parts', function() {
+        var l = lex('SUMMARY_PARTS.md');
+        assert.equal(l.parts.length, 3);
     });
 
     it('should allow lists separated by whitespace', function() {
         var l = lex('SUMMARY_WHITESPACE.md');
-        assert.equal(l.chapters.length, 5);
+        assert.equal(l.parts[0].articles.length, 5);
     });
 
     it('should allow ignore empty entries', function() {
         var l = lex('SUMMARY_EMPTY.md');
-        assert.equal(l.chapters.length, 1);
-    });
-
-    it('should throw error for sublist entries', function() {
-        assert.throws(
-            function() {
-                var l = lex('SUMMARY_SUBLIST.md');
-            },
-            "Invalid entry in the SUMMARY"
-        );
+        assert.equal(l.parts[0].articles.length, 1);
     });
 
     it('should correctly convert it to text', function() {